Abstact

SAR studies focused on improving the pharmacokinetic (PK) properties of the previously reported potent and selective Btk inhibitor CGI-1746 (1) resulted in the clinical candidate GDC-0834 (2), which retained the potency and selectivity of CGI-1746, but with much improved PK in preclinical animal models. Structure based design efforts drove this work as modifications to 1 were investigated at both the solvent exposed region as well as 'H3 binding pocket'. However, in vitro metabolic evaluation of 2 revealed a non CYP-mediated metabolic process that was more prevalent in human than preclinical species (mouse, rat, dog, cyno), leading to a high-level of uncertainly in predicting human pharmacokinetics. Due to its promising potency, selectivity, and preclinical efficacy, a single dose IND was filed and 2 was taken in to a single dose phase I trial in healthy volunteers to quickly evaluate the human pharmacokinetics. In human, 2 was found to be highly labile at the exo-cyclic amide bond that links the tetrahydrobenzothiophene moiety to the central aniline ring, resulting in insufficient parent drug exposure. This information informed the back-up program and discovery of improved inhibitors.